1

THE RAFFLES BULLETIN OF ZOOLOGY 2005

THE RAFFLES BULLETIN OF ZOOLOGY 2005 53(1): 1-11© National University of Singapore

PATTERNS OF POLYCHAETE COMMUNITIES IN TROPICAL SEDIMENTARYHABITATS: A CASE STUDY IN SOUTH-WESTERN THAILAND

Christopher R. S. Barrio Froján and Lawrence E. HawkinsSchool of Ocean and Earth Science, Southampton Oceanography Centre,

European Way, Southampton SO14 3ZH, UK

Chittima Aryuthaka and Sompoch NimsantijaroenDepartment of Marine Science, Faculty of Fisheries, Kasetsart University,

Bang Khen, Bangkok 10900, Thailand

Mike A. KendallPlymouth Marine Laboratory, Prospect Place, Plymouth PL1 3DH, UK

Gordon L. J. PatersonDepartment of Zoology, The Natural History Museum, Cromwell Road,

London SW7 5BD, UK

ABSTRACT. – Few studies have compared macrofaunal assemblages (organisms retained on a 0.5 mm sieve)between different tropical sedimentary habitats. The present study aims to address this issue, investigatingvegetated (seagrass), non-vegetated organic mud and exposed beach habitats. Polychaetes were the dominanttaxonomic group at all sampling sites, followed by crustaceans and molluscs. Thirty-six polychaete familieswere recorded, 31% of which were present in all three habitats. All families were represented in non-vegetatedhabitats. Polychaete family and functional group assemblages did not differ greatly between vegetated andnon-vegetated habitats despite varying densities. Differences in polychaete assemblage structure were mostpronounced between exposed beach and non-vegetated habitats. Polychaete assemblages were also negativelyaffected by the onset of the wet monsoon season. Overall, macrofaunal density (29 - 2,629 ind. m-2) fellwithin the range of that reported by other authors in the region. Small-scale physical and environmentalfactors are thought to be responsible for the large variation in polychaete assemblage structure within habitats.Identification of polychaetes to species, together with further analysis of small-scale physical data shouldprovide a deeper understanding of the spatial and temporal changes reported in this study. A call is madefor the standardisation of future sampling efforts and techniques to allow meaningful comparisons amongstsedimentary habitats across large spatial scales.

KEY WORDS. – seagrass, polychaete, macrofaunal density, inter-habitat.

INTRODUCTION

Tropical sedimentary habitats, though visually simple, arediverse and complex. They comprise a variety of substrata:from non-vegetated soft mud, sand and coarse shingle, to algalmats, seagrass beds and mangrove swamps, all sustained byvarying supplies of terrestrial- and marine-derived detritus.They may also be extremely dynamic (Alongi, 1990;Dittmann, 2000; Hall, 1994; Reise, 2002). Many studies (e.g.,Ellingsen & Gray, 2002; Kendall & Widdicombe, 1999; Zajacet al., 1998) have shown how environmental factors, such asvegetation, sedimentary characteristics and organicenrichment, affect benthic biodiversity at local scales.However, to address the growing interest in broad-scalepatterns of biodiversity (Gray, 2002; Kendall et al., 2003),

assessment needs explicitly-structured, comparably-collected data sets, which are largely lacking in the tropics.

The benthic biota of the Andaman Sea coast of Thailand ispoorly known in comparison with inshore benthic habitatsin Europe or North America but it has recently been thesubject of a number of scientific studies: Dexter (1996),Hylleberg & Nateewathana (1983; 1991a; 1991b),Meksumpun & Meksumpun (1999), Nateewathana (1988),and Petersen & Curtis (1980). Despite this body of work,there is still a lack of data sets comparable with those fromhigher latitudes. The present study builds on earlier workin the region and provides the much-needed, purpose-collected data, which will enable small and broad-scalecomparisons of benthic biodiversity.

2

Barrio Froján et al.: Polychaete communities in Thailand

To address all the issues concerning broad-scale patterns forthe fauna of the Andaman Sea is a substantial undertaking,but a first approximation can be achieved by bringing togetherexisting data with information collected during a series ofstudies undertaken from Ranong Coastal Resource ResearchStation. In this paper we will: (i) investigate the effects ofvegetation/habitat on polychaete assemblages in the region,(ii) identify the extent of seasonal variability in polychaeteabundance, and (iii) compare our data to those reported insimilar habitats throughout the tropics and beyond. This isthe first comparative study of benthos from differentsedimentary habitats to be carried out in Thailand.

MATERIALS AND METHODS

Site selection. – Collections have been made from anapproximately 50 km long stretch of the Andaman coast ofsouthwestern Thailand (Fig. 1) as part of a three-year baselinebiodiversity survey. The shoreline is characterised bymangrove forests, small estuaries with patchy seagrass bedsand long sandy beaches. Poorly developed fringing reefs arealso present around islands and rocky outcrops (Chansang &Poovachiranon, 1994). Eleven sites were sampled between

March 2000 and November 2001 and selected to representthree major sedimentary habitat types observed in the area:non-vegetated organic-rich mud, vegetated (seagrass)sediments, and exposed sandy beaches (Table 1). Distinctionsbetween these three habitats were made on visual and physicalcriteria.

Environmental regime of the study area. – There are twoseasons: the southwest monsoon season prevails betweenMay and October, and the northeast monsoon betweenNovember and April. The southwest monsoon is characterisedby heavy rainfall and strong onshore winds, which createcontinual moderate to high wave action. In contrast, duringthe northeast monsoon season, or dry season, this stretch ofcoastline experiences no significant rain, limited wind and areduced intensity, frequency and height of waves. Thesemidiurnal tidal amplitude ranges from 1.1 to 3.2 m with amean of about 2.2 m (Chansang & Poovachiranon, 1994).Five sites (of the 11) covering the three types of habitat weresampled during both the southwest (wet) monsoon and thenortheast (dry) monsoon (Table 1).

Quantitative sampling. – Sampling was carried out onexposed flats nearest the water’s edge at 11 selected locationsduring low tide using a 30 cm diameter (area: 0.07 m2) plastictube corer, inserted to a maximum depth of 40 cm. Adjacentsurface samples were taken for sediment granulometricanalysis. Granulometry samples were dried, sieved on a stackof graded sieves ranging from 63 µm to 2000 µm mesh, andthe residue on each weighed (Buchanan & Kain, 1971).Faunal samples were washed through a 500 µm gauge sieve.All sieving residue was fixed in 8 % formalin and latertransferred to 70 % alcohol before sorting under a dissectionmicroscope and identification of macrofauna. All animalswere identified to the lowest practical taxonomic level.Thailand has a rich polychaete fauna (Kendall et al., 2000),much of which cannot presently be fully identified as manyspecies remain undescribed. The lack of taxonomicinformation inhibits full species by species comparison ofall available data. Until more, fully comparable speciesinformation becomes available, initial investigation ofpatterns of polychaete distribution will be carried out at familylevel. Functional implications of patterns in faunal abundancewere examined by an analysis of feeding category, in whichfood preference, motility pattern and feeding structuremorphology are considered, as developed for polychaetefamilies by Fauchald & Jumars (1979).

Analytical procedures. – Differences in the density ofpolychaetes between habitats and sampling seasons wereexamined using the non-parametric Mann-Whitney (U) test.Spatial patterns in the distribution of the polychaetepopulation sampled were analysed using the PRIMERsoftware package (Clarke & Warwick, 1997). To detectpossible differences in polychaete assemblage composition(i.e., in the number and identity of families with or withoutthe influence of their relative abundance) between habitats,multivariate analyses were carried out on untransformed andtransformed family abundance data. Bray-Curtis index andgroup average linkage were used for cluster analysis and non-

Fig. 1. Map of study area showing the location of each samplingsite. Site abbreviations: AK = Ao Khoei, KKam = KhlongKamphuan, KKap = Khlong Kapoe, KKY = Ko Kai Yai, KKN = KoKam Nui, KRa = Ko Ra, KThao = Ko Thao, LSon = Laem Son,TN = Thale Nok, TND = Thung Nang Dam, TNDb = Thung NangDam beach.

3

THE RAFFLES BULLETIN OF ZOOLOGY 2005

Table 1. Physical information and sampling regime pertaining to each site sampled. Shaded months are those falling in the comparativelydryer northeast monsoon.

Number of samples taken

Position 2000 2001

Site Latitude (N) Longitude (E) Habitat % Sand Mar Sep Nov Mar Apr May Nov

AK 9.178 98.227 EB 72 15 3KKam 9.229 98.239 NV - 4 5 1KKap 9.346 98.296 SGHo, Hb - 2KKN 9.241 98.238 EB - 3KKY 9.201 98.195 NV 67 10KRa 9.157 98.185 NV 51 3 12KThao 9.577 98.491 NV - 6LSon 9.556 98.496 SGHb - 6TN 9.276 98.259 EB 90 6TND 9.135 98.201 SGHo, Cr 52 5 12TNDb 9.135 98.201 NV 62 2 1

Site abbreviations as in Fig. 1. Habitat abbreviations: EB = Exposed Beach; NV = Non-vegetated sediment; SG = Seagrass. % Sand refersto the combined proportion of sediment retained on 250 µm and 125 µm mesh-size sieves; no values appear where granulometry data hasbeen lost. Ho = Halophila ovalis, Hb = Halophila beccarii, Cr = Cymodocea rotundata.

metric multidimensional scaling (MDS) ordination. One-wayANOSIM (analysis of similarities) was carried out to testdifferences found in polychaete communities betweenhabitats, and families typifying each habitat were identifiedusing the SIMPER (similarity percentages) programme.

RESULTS

Granulometry analysis of the study sites indicated littledifferentiation in particle size ratios between seagrass, non-vegetated and exposed beach habitats, all falling within the‘muddy gravelly sand’ category defined in the HabitatIdentification Manual for Ranong (Kendall et al., in prep.).

Macrofauna. – A total of 6,774 macrofaunal individuals werecollected in 96 cores at all 11 sites (Appendix). Polychaeteworms were numerically dominant, accounting for over 75%of the total macrofauna sampled. Crustacea and Molluscafollowed, together constituting around 14% of allmacrofauna. Mean density of macrofaunal organismscore-1 was (± SE) 68.4 ± 17.1 (n = 91). There was a highdegree of variability in mean macrofaunal density betweensites, ranging from 2.0 ± 0.0 individuals core-1 (n = 3) at KoKam Nui (KKN) to 184.0 ± 35.0 individuals core-1 (n = 6)at Laem Son (LSon). Macrofauna were present at a higherdensity (Fig. 2) in seagrass (1,614 ± 571 m-2) and in non-vegetated habitats (1,127 ± 211 m-2) than in exposed beachhabitats (90 ± 44 m-2) (Mann-Whitney test: z > P0.01).Differences in mean macrofaunal density values between non-vegetated and seagrass habitats were not significant. Thepolychaete component of the macrofauna mirrored theserelationships (Fig. 2).

Polychaeta. – A total of 5,260 individuals belonging to 36families were collected (Table 2). Numerically, familiesCapitellidae, Paraonidae, Nereididae, Spionidae, Goniadidaeand Orbiniidae dominated in that order. Together theyconstituted over 64% of the individuals. Most other families

Fig. 2. Mean density (± SE) for polychaetes and macrofauna ateach sampled habitat. EB = Exposed Beach, NV = Non-vegetatedsediment, SG = Seagrass.

were less common, with approximately one quarter of thetotal number of families encountered being represented byless than 10 individuals. Orbiniidae was the most widespreadfamily, present at all 11 sites sampled, followed by theCapitellidae, Spionidae, Lumbrineridae, which were presentat 10 of the 11 sites. Cossuridae, Dorvilleidae, Onuphidaeand Pholoidae were present only at non-vegetated sites,which, as a habitat, contained all 36 families encountered.Non-vegetated sites harboured a greater number of rarefamilies than did seagrass sites. Seagrass sites contained 30families whilst exposed beach sites had only 12 families, noneexclusive to any habitat. Eleven families (31% of the total)were found in all three habitats. Overall, there is a generalincrease in the number of families sampled with the number

4

Barrio Froján et al.: Polychaete communities in Thailand

of replicate cores taken (Fig. 3) but the numbers of familiesappear to be highest at Ko Khai Yai (KKY) (n = 10) and KoRa (KRa) (n = 15) than at other sites. There is not, however,a significant trend of increasing polychaete density with anincrease in the number of replicate cores taken, suggestingthat the area sampled during the present study is notfaunistically under-represented. The difference inpolychaete density between monsoon seasons, thoughvariable from habitat to habitat, is significant (Mann-Whitneytest: z >> P0.01), being higher in the dry season than in the wetmonsoon season when the whole area under investigation isconsidered (Fig. 4). This trend was reversed in seagrasshabitats.

The functional composition of the polychaete fauna of thestudy area has been examined in terms of: (i) the mostabundant families within each habitat (i.e., those contributingcumulatively > 40% of individuals in their habitat), (ii) thefamilies accounting for 95% of all individuals within eachhabitat, and (iii) the total number of families present in eachhabitat (Table 3). When only the most abundant families areexamined (group (i)), all feeding categories except filterfeeders were equally represented. When groups (ii) and (iii)are considered, carnivores were the most represented feedingcategory, followed by surface and sub-surface deposit feeders,in that order. Differences in the abundance of each feedingcategory between seagrass and non-vegetated habitats werenot significant (Table 3). Significant differences in therelative abundance of the different feeding categories weredetected between exposed beach and the other two habitatsfor all but the herbivore and filter feeding polychaete families.

Analysis of community structure. – Polychaetes, the mostabundant and well-characterised component of sampledmacrofauna, have been used exclusively in the followinginterhabitat comparisons. Tests for differences in polychaetecommunity structure – performed using multivariate analysison untransformed data, thus taking into account relativefamily abundance – indicated an overall lack of dissimilaritybetween habitats (ANOSIM result global R statistic: 0.292,P = 0.05), i.e., there were no distinct polychaete family

Fig. 3. Relationships between family richness (Margalef’s d = (S-1)/Log(N)) (�), density (�) and sample replicates at the 11 sitessampled. Site abbreviations as in Fig. 1.

Fig. 4. Box and whisker plot of seasonal variation at each habitatand for all habitats considered together. Left-hand side boxesrepresent dry season sampling; right-hand side boxes represent wetseason sampling. Habitat abbreviations as in Fig. 2.

assemblages in any habitat. Pairwise comparisons betweenhabitats were equally indistinct. Only with severetransformation of the data, i.e., by only assigning familiespresence/absence (+/-) values, thus eliminating the effect ofrelative abundance and relying on similarities in polychaetefamily diversity, did any significant difference appear (Table4), yet only between non-vegetated and exposed beachhabitats (ANOSIM result for pairwise EB-NV +/- comparisonR statistic: 0.764, P < 0.02). No differences in polychaetefamily diversity were found between non-vegetated andvegetated habitats. The relationships here highlighted areillustrated with the MDS plots in Figs. 5a and 5b, whereresolution between vegetated and non-vegetated habitats islacking, even when exposed beach habitats are omitted fromthe analysis.

This pattern of overall similarity amongst habitats yet weakdifferences in pairwise comparisons between exposed beachand non-vegetated habitats is also reflected when usingpolychaete feeding group abundance figures (Table 4). Therelative contribution of families and feeding categoriestypifying each habitat was identified using SIMPER analysis(Table 5). The variation in proportion of sub-surface depositfeeders to carnivores was responsible for the differentiationbetween the two most dissimilar habitats.

DISCUSSION

Ecological comparisons. – In-depth analysis of communitystructure and function has revealed subtle patterns of variationbetween the habitats under investigation.

Previous studies (e.g., Turner & Kendall, 1999; Somaschiniet al., 1994; Edgar et al., 1994; Ansari et al., 1991; Sheridan,1997) have all – using slightly differing methodology andanalyses – reported a greater diversity and abundance ofinfauna in vegetated than in adjacent non-vegetated areas.Most attribute this difference quantitatively to habitatmodification by emergent seagrass plants altering localbiological, chemical and/or physical conditions, which in turn

5

THE RAFFLES BULLETIN OF ZOOLOGY 2005

Table 2. Total numbers of individuals belonging to major macrobenthic taxa at each habitat sampled on the north-western coast of southernThailand. Polychaete families are followed by a three-letter feeding category code.

Habitat

SG NV EB Σ

Replicate cores (n) 20 (25) 44 27 91 (96)Polychaeta 1,646 (2,323) 2,721 216 4,583 (5,260)

Ampharetidae sst 69 22 0 91Amphinomidae cmx 83 55 0 138Capitellidae bmx 305 460 32 797Chaetopteridae fsp 1 1 0 2Chrysopetalidae cmx 9 5 0 14Cirratulidae smt 40 152 1 193Cossuridae bmx 0 8 0 8Dorvilleidae smj 0 2 0 2Eunicidae cmj 14 17 0 31Flabelligeridae sdt 4 1 0 5Glyceridae cdj 43 60 18 121Goniadidae cdj 218 124 0 342Hesionidae hmj 6 2 0 8Lumbrineridae cmj 49 113 8 170Magelonidae sdt 16 52 0 68Maldanidae bsx 65 35 0 100Nephtyidae cmj 52 60 0 112Nereididae cmj 293 379 2 674Oenonidae cmj 2 1 0 3Onuphidae cmj 0 6 0 6Opheliidae bmx 9 21 2 32Orbiniidae bmx 169 104 32 305Oweniidae sdt 91 69 0 160Paraonidae hmx 163 440 91 694Pholoidae cmj 0 3 0 3Phyllodocidae cmx 12 26 1 39Pilargidae cmj 33 108 0 141Poecilochaetidae sdt 6 11 0 17Polynoidae cmj 8 3 0 11Sabellidae fst 62 11 0 73Spionidae sdt 333 236 25 594Sternapsidae bmx 13 44 0 57Syllidae cmj 131 68 1 200Terebellidae sst 3 9 0 12Trichobranchidae sst 21 2 0 23Unknown 0 11 3 14

Crustacea 347 302 27 676Mollusca 326 51 6 383Sipuncula 148 93 0 241Echinodermata 20 10 4 34Other taxa 37 141 2 180Total 2,524 (3,201) 3,318 255 6,097 (6,774)Polychaety density (m-2) 1,327 883 114 783Macrofauna density (m-2) 1,803 1,077 135 957

Feeding category three-letter code: first letter (major food) – b = sub-surface deposit feeder, c = carnivore, f = filter feeder, h = herbivore,s = surface deposit feeder; second letter (motility) – d = discreetly motile, m = motile, s = sessile; third letter (feeding structure) – j = jaws,p = pump, t = tentacles, x = other, such as eversible pharynges (Fauchald & Jumars, 1979). ‘Other taxa’ include Enteropneusta, Anthozoa,Turbellaria, Cephalochordata, small vertebrates and insect larvae. Numbers in parentheses indicate the addition of five replicate samplesfor which there are only polychaete data; numbers outside parentheses have been used for macrofaunal density calculations. Habitatabbreviations as in Table 1.

6

Barrio Froján et al.: Polychaete communities in Thailand

Table 3. Distribution of families by feeding category and habitat, related to the number of families examined. Habitat and feeding categoryabbreviations as in Tables 1 & 2, respectively.

Most abundant families 95% of families All families

Feeding category SG EB NV SG EB NV SG EB NV

Sub-surface deposit feeders bmx 1 0 1 2 2 3 4 3 5bsx 0 0 0 1 0 1 1 0 1

Carnivores cdj 0 0 0 2 1 2 2 1 2cmj 1 0 1 5 1 5 8 3 10cmx 0 0 0 1 0 2 3 1 3

Filter feeders fsp 0 0 0 0 0 0 1 0 1fst 0 0 0 1 0 0 1 0 1

Herbivores hmj 0 0 0 0 0 0 1 0 1hmx 0 1 1 1 1 1 1 1 1

Surface deposit feeders sdt 1 0 0 2 1 3 5 1 5smj 0 0 0 0 0 0 0 0 1smt 0 0 0 1 0 1 1 1 1sst 0 0 0 1 0 0 3 0 3

All categories 3 1 3 17 6 18 31 11 35

Interhabitat comparison of the numbers of families within each polychaete feeding category.

SG-NV NV-EB EB-SG

Sub-surface deposit feeders n.s. * *Carnivores n.s. * *Filter feeders n.s. n.s. n.s.Herbivores n.s. n.s. n.s.Surface deposit feeders n.s. * *

Most abundant families are those contributing cumulatively > 40 % of individuals within their habitat. Lord’s Range test used where n <5 (i.e. for Filter feeders and Herbivores). Asterisk (*) indicates a significant difference (z or L ≥ P(0.01)), n.s. = not significant.

Table 4. PRIMER output summary for ANOSIM tests carried out on presence/absence transformed polychaete family data and feedingcategory abundance at each habitat. Habitat abbreviations as in Table 1.

Family (+/-) Feeding category

R statistic Sig. level (%) R statistic Sig. level (%)

Global 0.405 1.8 0.304 3.3Pairwise comparisonsEB-NV 0.764 1.8 0.631 1.8EB-SG 0.556 10.0 0.259 30.0NV-SG -0.041 57.1 0.067 37.5

Table 5. Polychaete families and feeding categories typifying each habitat resulting from SIMPER analysis ranked in decreasing order oftheir importance to the similarity within their habitat. Figures alongside each entry indicate total % contribution. Only families cumulativelycontributing c. 50 % to the similarity are listed. Habitat abbreviations as in Table 1.

SG NV EB

Family Feeding cat. Family Feeding cat. Family Feeding cat.

Spionidae 19.6 carnivore 46.9 Capitellidae 30.2 carnivore 40.1 Orbiniidae 74.1 ss-deposit 71.0Capitellidae 19.2 ss-deposit 24.0 Paraonidae 15.2 ss-deposit 33.6 carnivore 17.0Nereididae 11.0 s-deposit 24.0 Spionidae 9.9 s-deposit 15.8 s-deposit 11.0

herbivore 2.1 herbivore 10.0filter 1.0 filter 0.5

7

THE RAFFLES BULLETIN OF ZOOLOGY 2005

Fig. 5. Multivariate analysis output using presence/absencetransformed polychaete data: (a) MDS plot showing all sitesarranged by habitat, (b) MDS plot excluding Exposed Beach habitat.Habitat abbreviations as in Fig. 2.

(b)

(a)

enhance deposition of material in suspension (e.g., finesediments, detritus, plankton and invertebrate larvae) withconcomitant effects on the structure of benthic communities(Eckman, 1983). Differences in macrofaunal density betweenseagrass and non-vegetated habitats in the present study maynot be supported statistically, yet the sheer variability ofmacrofaunal density figures between cores at seagrass sitesreflects the increased heterogeneity and complexity of thishabitat when compared with non-vegetated habitats. It is thishigh degree of variation in macrofaunal density within andacross sites vegetated with differing seagrass species yetbelonging to the same habitat that has likely reduced the levelof significance of the difference between habitats. Datatransformation and allocation of polychaete families intofunctional groups reduced variability and revealed strongerand slightly significant relationships between habitats.

Habitat complexity is compounded from quantitative (thevariety in types of substrata) and qualitative (the amount ofabove-ground substratum) elements. Qualitative complexitycan be contributed in the form of different plant architectures,food sources, or habitat topography and attachment sites. Inthe present study it was found that at small spatial scales (<1 m), neither seagrass nor non-vegetated sites werefaunistically or topographically homogeneous, both beinggreatly influenced by intensive small-scale bioturbation (e.g.,by burrowing sipunculans, crabs and shrimp), thus

augmenting within-habitat physical variability. The negativeR statistic resulting from the ANOSIM analysis between thesehabitats (Table 4) is indicative of greater variation betweensamples within a habitat than between habitats (Chapman &Underwood, 1999), thus supporting this observation. It hasbeen assumed that at small scales there is a relationshipbetween faunal similarity and the separation of any pair ofsamples but there is no basis for such an assumption (Kendall& Widdicombe, 1999). Macrofauna were found not to berandomly distributed within a habitat but were reflecting thissmall-scale spatial patchiness. Identification of polychaetesto a lower taxonomic level and subsequent analysis may alterthe significance of this relationship and work is presentlyunderway to investigate this.

Several authors report negative effects of monsoon rain onintertidal benthic fauna (see review by Alongi, 1990), listingincreased levels of sediment disturbance by stronger waveaction as the major responsible factor. This study does notcontradict these findings as regards changes in overallpolychaete density between seasons. However, variation ofresponse to changes in season between habitats is, at best,erratic (Fig. 4). Vegetated sites, contrary to expectation, hada higher abundance of macrofauna during the wet monsoon.This may be due to seagrass fronds buffering most of thepotentially damaging wave action and simultaneouslytrapping more of the increased suspended particulate matterpresent in the water column. Since all granulometry valuesfell within the same sediment category regardless of habitator time of sampling, further investigation is needed beforeany process affecting this particular seasonal change can beidentified.

Sheridan (1997) related benthic macrofaunal abundance tocommunity function by allocating taxa into feeding groups.He found that the total faunal community in southwestFlorida’s intertidal substrates did not show major differencesin feeding group composition between relatively stablesedimentary habitats (mangrove, seagrass and non-vegetated)despite these habitats differing in species composition.Similarly, in the present study, no great differences wereobserved between seagrass and non-vegetated habitats.Carnivores were most abundant in both these habitats,followed in turn by sub-surface and surface deposit feeders(Table 5). This pattern of distribution of trophic categoriespoints to an established community with an evolved foodchain reflecting a relatively stable environment in thesehabitats. Exposed sandy beach habitats, on the other hand,had an exclusive and overwhelming dominance of pioneeringand opportunistic motile sub-surface deposit feeders (Table5), thus reflecting a more dynamic, less stable environment,refractory to the settlement of a diverse and maturecommunity. Interestingly, in the present study only non-vegetated habitats had a significant proportion (> 10%) ofrare and herbivorous families. This could reduce the relativeimportance of vegetation as a stabilising influence of benthicmacro-infaunal communities, as well as strengthening therelative value of non-vegetated habitats against vegetatedhabitats when considering their importance in coastal zonemanagement and conservation. Conservation bodies targeting

8

Barrio Froján et al.: Polychaete communities in Thailand

non-vegetated habitats containing greater numbers of rarefamilies may ensure a bigger representation and protectionof infaunal diversity. As the most extensive habitat aroundthe coast in the region, efforts to conserve small parts of itmay not encounter as much opposition as other, morecontentious and productive vegetated habitats.

Geographical comparisons. – Aungtonya et al. (2002) list atotal of 37 polychaete families known to occur in Thai waters.Representatives of eight of those families (Acoetidae,Eulepethidae, Pectinariidae, Pisionidae, Sabellariidae,Serpulidae, Sigalionidae and Trochochaetidae) have not beenencountered in the present study; however, it has generatedfive additional families to those already collected. These areCossuridae, Oenonidae, Paraonidae, Pholoidae andPhyllodocidae. It would thus appear that no fewer than 42polychaete families have been recorded in Thai waters, 78%of the total of 54 families reported to occur in and around theSouth China Sea (Paxton & Chou, 2000).

Total macrofaunal density for the study area (29 - 2,629ind. m-2) was within the broad limits of macrofaunal densityfound elsewhere in the Indo-Pacific region (Table 6).However, much higher and more variable densities have beenfound in other tropical settings, as well as in the temperatecoastal waters of Europe (e.g., Junoy & Vieitez, 1992; Reiseet al., 1994; Turner & Kendall, 1999), southern Africa (e.g.,Allanson et al., 2000; Bursey & Wooldridge, 2002) and NorthAmerica (e.g., Mattila et al., 1999). Differing definitions offaunal size categories prevent meaningful comparisonsbetween all authors and locations; however, works referredto in Table 6 are comparable with the present study and givean indication of the density of polychaetes at their respectivelocations. Values obtained in this study fall within the rangesrecorded by authors quoted in Table 6, yet most closelyresemble figures presented by Angsupanich & Kuwabara(1995): 630 - 1,121 ind. m-2, and Angsupanich & Kuwabara(1999): 120 - 575 ind. m-2 in Thai waters.

Reise (1991) reports a trend of decreasing macrofaunaldensity with decreasing latitude and supports his findings witha brief literature survey supporting his conclusions. Tropicalstudies from which he quotes low macrofaunal densities are,however, limited to mangrove swamps, thus omitting other

varied and highly variable tropical intertidal habitats.Moreover, Sheridan (1997) reports total macrofaunal densitiesto be highest in mangrove peat (22,591 - 52,914 ind. m-2)when compared with seagrass (6,347 - 23,545 ind. m-2) andmudflat (3,611 - 22,465 ind. m-2) sites. All of these tropicaldata are broadly similar to those reported for temperate sitesfor fauna retained on a 0.5 mm gauge sieve (Table 6). Sincesampling and sample-processing techniques are so variableamongst researchers, temporal variation in faunal abundanceso volatile, and habitat heterogeneity so pronounced atdifferent locations over small spatial scales, extrapolation ofmacrofaunal density from individuals core-1 to individualsm-2 can be misleading. A more standardised approach tobenthic sampling across the scientific community is requiredbefore any generalisation can be made about an area understudy. Only when fully representative and comparablycollected data sets are available can meaningful relationshipsbetween areas be sought.

ACKNOWLEDGEMENTS

This work was partly supported by a European Union fundedCapacity Building Project (Grant number THA/B76200/1B/1999/0342), a Faculty of Science scholarship from theUniversity of Southampton and The Natural History Museum,London. The authors are very grateful for all the help, researchfacilities and hospitality provided by all staff at the RanongCoastal Resource Research Station. Dave Parry and SteveWiddicombe helped in the sampling design, collection andprocessing of samples. Map courtesy of Jim Chimonides.Comments from the two anonymous referees helpedimproving the clarity of the manuscript and are gratefullyacknowledged.

LITERATURE CITED

Allanson, B. R., J. Nettleton, & C. J. De Villiers, 2000. Benthicmacrofauna richness and diversity in the Knysna Estuary: a 50year Comparison. Transactions of the Royal Society of SouthAfrica, 55: 177-185.

Alongi, D. M., 1990. The ecology of tropical soft-bottom benthicecosystems. Oceanography and Marine Biology: an AnnualReview, 28: 381-496.

Table 6. Comparison of intertidal macrofaunal (as retained on a 500 µm sieve) density from different locations.

Location Density (ind. m-2) Reference

TropicalColeroon Estuary, Bay of Bengal 50 - 2,500 Patterson Edward & Ayyakkannu (1992)Thale Sap Songkhla, Gulf of Thailand 374 - 2,420 Angsupanich & Kuwabara (1995; 1999)Mlonggo Bay, Java 463 - 1,069 Warwick & Ruswahyuni (1987)Bowling Green Bay, Australia 113 - 3,898 Dittmann (2000)Crooked Harbour, Hong Kong 3,200 - 40,700 Lee et al. (2001)Ala Wai canal, Hawaii 4,910 - 47,430 McCarthy et al. (2000)Jaltepeque Estuary, El Salvador 259 - 23,852 Lara & Zamora (1994)TemperateIsles of Scilly, UK 7,194 - 26,706 Bowden et al. (2001)Cellars Beach, UK 16,616 - 71,141 Turner & Kendall (1999)Yealm Estuary, UK 1,911 - 12,229 Webster et al. (1998)

9

THE RAFFLES BULLETIN OF ZOOLOGY 2005

Angsupanich, S., & R. Kuwabara, 1995. Macrobenthic fauna inThale Sap Songkla, a brackish lake in southern Thailand. Lakes& Reservoirs: Reserves and Management, 1: 115-125.

Angsupanich, S., & R. Kuwabara, 1999. Distribution ofmacrobenthic fauna in Phawong and U-Taphao canals flowinginto a lagoonal lake, Sonhkgla, Thailand. Lakes & Resevoirs:Reserves and Management, 4: 1-13.

Ansari, Z. A., C. U. Rivonker, P. Ramani, & A. H. Parulekar, 1991.Seagrass habitat complexity and macroinvertebrate abundancein Lakshadweep Coral-Reef Lagoons, Arabian Sea. Coral Reefs,10: 127-131.

Aungtonya, C., S. Thaipal, & S. Bussarawit, 2002. A list ofpolychaetes (Annelida) in the reference collection database ofthe Phuket Marine Biological Center, Thailand. In: Eibye-Jacobsen, D., (ed.), Proceedings of the International Workshopon the Polychaetes of the Andaman Sea. Phuket MarineBiological Center Special Publication, Phuket, Thailand. Pp.21-32.

Bowden, D. A., A. A. Rowden & M. J. Attrill, 2001. Effect of patchsize and in-patch location on the infaunal macroinvertebrateassemblages of Zostera marina seagrass beds. Journal ofExperimental Marine Biology and Ecology, 259: 133-154.

Buchanan, J. B. & M. Kain, 1971. Measurement of the physicaland chemical environment. In: McIntyre, A.D., (ed.), Methodsfor the study of marine benthos. Blackwell ScientificPublications, Oxford & Edinburgh. Pp 30-58

Bursey, M. & T. Wooldridge, 2002. Diversity of benthic macrofaunaof the flood-tidal delta of the Nahoon estuary and adjacent beach,South Africa. African Zoology, 37: 231-246.

Chansang H. & S. Poovachiranon, 1994. The distribution and speciescomposition of seagrass beds along the Andaman Sea coast ofThailand. Phuket Marine Biological Centre Research Bulletin,59: 43-52.

Chapman, M. G. & A. J. Underwood, 1999. Ecological patterns inmultivariate assemblages: information and interpretation ofnegative values in ANOSIM tests. Marine Ecology ProgressSeries, 180: 257-265.

Dexter, D. M., 1996. Tropical sandy beach communities of PhuketIsland, Thailand. Phuket Marine Biological Centre ResearchBulletin, 61: 1-28.

Dittmann, S., 2000. Zonation of benthic communities in a tropicaltidal flat of northeast Australia. Journal of Sea Research, 43:33-51.

Eckman, J. E., 1983. Hydrodynamic processes affecting benthicrecruitment. Limnology and Oceanography, 28: 241-257.

Edgar, G. J., C. Shaw, G. F. Watsona & L. S. Hammond, 1994.Comparisons of species richness, size-structure and productionof benthos in vegetated and unvegetated habitats in WesternPort, Victoria. Journal of Experimental Marine Biology andEcology, 176: 201-226.

Ellingsen, K. E. & J. S. Gray, 2002. Spatial Patterns of benthicdiversity: is there a latitudinal gradient along the Norwegiancontinental shelf? Journal of Animal Ecology, 71: 373-389.

Fauchald, K. & P. A. Jumars, 1979. The diet of worms: a study ofpolychaete feeding guilds. Oceanography and Marine Biology:an Annual Review, 17: 193-284.

Gray, J. S., 2002. Species richness of marine soft sediments. MarineEcology Progress Series, 244: 285-297.

Hall, S. J., 1994. Physical disturbance and marine benthiccommunities - life in unconsolidated sediments. Oceanographyand Marine Biology: an Annual Review, 32: 179-239.

Hylleberg, J. & A. Nateewathana, 1983. Responses of polychaetefamilies to monsoon- and offshore mining-associated sedimentdisturbance. In: Hutchings, P.A., (ed.), Proceedings of the FirstInternational Polychaete Conference. The Linnean Society ofNew South Wales, Sydney, Australia. Pp. 279-291.

Hylleberg, J. & A. Nateewathana, 1991a. Spatial and temporaldistribution of spionid polychaetes at Phuket Island, theAndaman Sea. Bulletin of Marine Science, 48: 346-357.

Hylleberg, J. & A. Nateewathana, 1991b. Temporal and spatialdistribution of subtidal magelonid polychaetes at Phuket Island,Andaman Sea, Thailand. Ophelia Supplement, 5: 573-578.

Junoy, J. & J. M. Vieitez, 1992. Macrofaunal abundance analysesin the Ria de Foz (Lugo, northwest Spain). Cahiers De BiologieMarine, 33: 331-345.

Kendall, M. A., G. L. J. Paterson, S. Nimsantijaroen, in prep. AGuide to Marine Habitats of Ranong.

Kendall, M. A., G. L. J. Paterson & C. Aryuthaka, 2000. On-lineexchange of polychaete taxonomic information. Bulletin ofMarine Science, 67: 411-420.

Kendall, M. A. & S. Widdicombe, 1999. Small scale patterns in thestructure of macrofaunal assemblages of shallow soft sediments.Journal of Experimental Marine Biology and Ecology, 237: 127-140.

Kendall, M. A., S. Widdicombe & J. M. Weslawski, 2003. A multi-scale study of the biodiversity of the benthic infauna of the high-latitude Kongsfjord, Svalbard. Polar Biology, 26: 383-388.

Lara, O. A. M. & J. A.V. Zamora, 1994. Structure of themacrobenthos in Jaltepeque estuary, El Salvador. Revista DeBiología Tropical, 42: 165-174.

Lee, S. Y., C. W. Fong & R. S. S. Wu, 2001. The effects of seagrass(Zostera japonica) canopy structure on associated fauna: a studyusing artificial seagrass units and sampling of natural beds.Journal of Experimental Marine Biology and Ecology, 259: 23-50.

Mattila, J., G. Chaplin, M. R. Eilers, K. L. Heck, J. P. O’Neal &J. F. Valentine, 1999. Spatial and diurnal distribution ofinvertebrate and fish fauna of a Zostera marina bed and nearbyunvegetated sediments in Damariscotta River, Maine (USA).Journal of Sea Research, 41: 321-332.

McCarthy, S. A., E. A. Laws, W. A. Estabrooks, J. H. Bailey-Brock& E. A. Kay, 2000. Intra-annual variability in Hawaiian shallow-water, soft-bottom macrobenthic communities adjacent to aeutrophic estuary. Estuarine, Coastal and Shelf Science, 50: 245-258.

Meksumpun, C. & S. Meksumpun, 1999. Polychaete-sedimentrelations in Rayong, Thailand. Environmental Pollution, 105:447-456.

Nateewathana, A., 1988. Heptaceras hyllebergi n. sp. (Polychaeta:Onuphidae) from Kata beach, west coast of Phuket Island,Andaman Sea, Thailand, with notes on the habitat. PhuketMarine Biological Center Research Bulletin, 48: 1-9.

Patterson Edward, J. K. & K. Ayyakkannu, 1992. Benthicmacrofauna of Coleroon estuary, southeast coast of India.Phucket Marine Biological Centre Research Bulletin, 57: 67-76.

Paxton, H. & L. M. Chou, 2000. Polychaetous annelids from theSouth China Sea. The Raffles Bulletin of Zoology, 8: 209-232.

Petersen, G. H. & M. A. Curtis, 1980. Differences in energy flowthrough major components of subarctic, temperate and tropicalmarine shelf ecosystems. Dana, 1: 53-64.

10

Barrio Froján et al.: Polychaete communities in Thailand

Reise, K., 1991. Macrofauna in mud and sand of tropical andtemperate tidal flats. In: Elliott, M. & J. P. Ducrotoy, (eds.),1991. Estuaries and Coasts: Spatial and TemporalIntercomparisons. Olsen and Olsen, Fredensborg. Pp. 211-216.

Reise, K., 2002. Sediment mediated species interactions in coastalwaters. Journal of Sea Research, 48: 127-141.

Reise, K., E. Herre, & M. Sturm, 1994. Biomass and abundance ofmacrofauna in intertidal sediments of Konigshafen in thenorthern Wadden Sea. Helgolander Meeresuntersuchungen, 48:201-215.

Sheridan, P., 1997. Benthos of adjacent mangrove, seagrass and non-vegetated habitats in Rookery Bay, Florida, USA. Estuarine,Coastal and Shelf Science, 44: 455-469.

Somaschini, A., M. F. Gravina & G. D. Ardizzone, 1994. Polychaetedepth distribution in a Posidonia oceanica bed (rhizome andmatte strata) and neighboring soft and hard bottoms. Marine

Ecology - Pubblicazioni della Stazione Zoologica di Napoli I,15: 133-151.

Turner, S. J. & M. A. Kendall, 1999. A comparison of vegetatedand unvegetated soft-sediment macrobenthic communities in theriver Yealm, southwestern Britain. Journal of the MarineBiological Association of the United Kingdom, 79: 741-743.

Warwick, R. M. & Ruswahyuni, 1987. Comparative-study of thestructure of some tropical and temperate marine soft-bottommacrobenthic communities. Marine Biology, 95: 641-649.

Webster, P. J., A. A. Rowden & M. J. Attrill, 1998. Effect of shootdensity on the infaunal macro-invertebrate community withina Zostera marina seagrass bed. Estuarine, Coastal and ShelfScience, 47: 351-357.

Zajac, R. N., R. B. Whitlatch & S. F. Thrush, 1998. Recolonizationand succession in soft-sediment infaunal communities: thespatial scale of controlling factors. Hydrobiologia, 376: 227-240.

11

THE RAFFLES BULLETIN OF ZOOLOGY 2005

Appendix. Total macrofauna sampled during March 2000 – November 2001 in 96 cores over 11 sites.

AK KKam KKap KKY KKN KRa KThao LSon TN TND TNDb Σ

PolychaetaCapitellidae 28 194 13 36 0 143 38 79 4 213 49 797Cossuridae 0 0 0 0 0 0 8 0 0 0 0 8Opheliidae 2 0 0 10 0 9 2 1 0 8 0 32Orbiniidae 15 12 2 18 3 49 5 14 14 153 20 305Sternapsidae 0 0 13 0 0 40 4 0 0 0 0 57Maldanidae 0 3 3 15 0 8 1 26 0 36 8 100Glyceridae 17 0 1 6 1 33 21 34 0 8 0 121Goniadidae 0 6 11 2 0 86 6 9 0 198 24 342Eunicidae 0 0 0 2 0 7 0 1 0 13 8 31Lumbrineridae 7 5 1 61 1 27 12 1 0 47 8 170Nephtyidae 0 2 0 7 0 50 1 0 0 52 0 112Nereididae 2 6 7 16 0 29 326 239 0 47 2 674Oenonidae 0 0 0 0 0 1 0 1 0 1 0 3Onuphidae 0 0 0 5 0 0 0 0 0 0 1 6Pholoidae 0 0 0 0 0 3 0 0 0 0 0 3Pilargidae 0 57 0 2 0 40 9 24 0 9 0 141Polynoidae 0 0 0 2 0 1 0 2 0 6 0 11Syllidae 1 0 0 32 0 18 1 55 0 76 17 200Amphinomidae 0 0 0 1 0 26 28 26 0 57 0 138Chrysopetalidae 0 0 1 5 0 0 0 7 0 1 0 14Phyllodocidae 0 4 0 3 0 14 5 4 1 8 0 39Chaetopteridae 0 0 0 0 0 1 0 0 0 1 0 2Sabellidae 0 0 0 5 0 1 2 0 0 62 3 73Hesionidae 0 0 3 0 0 2 0 0 0 3 0 8Paraonidae 91 2 4 70 0 285 23 27 0 132 60 694Flabelligeridae 0 0 0 1 0 0 0 0 0 4 0 5Magelonidae 0 2 10 34 0 12 4 0 0 6 0 68Oweniidae 0 0 5 34 0 7 11 7 0 79 17 160Poecilochaetidae 0 0 2 1 0 8 2 4 0 0 0 17Spionidae 18 14 5 57 0 131 26 116 7 212 8 594Dorvilleidae 0 0 0 2 0 0 0 0 0 0 0 2Cirratulidae 1 0 4 20 0 120 8 6 0 30 4 193Ampharetidae 0 2 0 0 0 11 6 6 0 63 3 91Terebellidae 0 0 0 2 0 3 2 0 0 3 2 12Trichobranchidae 0 0 0 0 0 2 0 0 0 21 0 23Unknown 3 0 0 2 0 7 2 0 0 0 0 14

Crustacea 27 13 1 42 0 185 35 75 0 271 27 676Mollusca 5 0 0 9 1 16 22 295 0 31 4 383Sipuncula 0 3 0 3 0 37 37 23 0 125 13 241Echinodermata 4 0 1 2 0 6 1 6 0 13 1 34Other taxa 0 5 2 50 0 60 14 10 2 25 12 180

Σ 221 330 89 557 6 1,478 656 1,104 28 1,337 291 6,774

Macrofaunal data for TND are only available for 12 replicate cores. ‘Other taxa’ include Enteropneusta, Anthozoa, Turbellaria, Cephalochordata,small vertebrates and insect larvae. Site abbreviations as in Fig. 1.